Latest Info for Wide Gamut w/ Questionable HW Corrections – BenQ SW271

[Gette]

Hi…

First thank you so much for help in advance as I have been spending a few days researching, reading and testing trying to get this BenQ SW271 dialed in as best as possible.

I am getting stuck on a few things and not sure what is the best way to approach it, as there seems to be a few recommended ways (though some may be dated) and you guys have made updates since then.

Benq SW271 via Display Port:
https://www.displayspecifications.com/en/model/5be2cec

Mac Pro 5,1 – High Sierra (soon to be Mojave) – Nvidia GTX 1070
(I also have a Radeon RX580 if that is preferred for whatever reason).
Mainly using for Adobe Lightroom/Photoshop (photography)
Xrite i1 Display Pro

Seemingly the recommendation is to have/use built in HW calibration through BenQ’s version of Palette Master Elements. Though it sounds like that may not be using the correct correction so I am not sure the best place to start if that is in fact 100% correct.

Questions:
1 – If that’s the case is it still best to do/try a custom PME HW calibration with Native Color Space (65k, G 2.2, Relative, 16 Lut, etc) and then DisplayCal Profile on top (or any other setting  recommendations)?
Or use their Factory AdobeRGB preset and do a DisplayCal Profile on top of that or (as opposed to creating my own is that will be more accurate then mine esp if their Correction is not correct)?

2 – What DisplayCal Correction is the best/latest correct one?
I have tried both the built in:
“LCD PFS Phosphor WLED IPS, 98% Adobe RGB…HP Z24x…”
and also saw a few in the Database:
https://colorimetercorrections.displaycal.net/?get&type=ccss&manufacturer_id=BNQ&display=BenQ%20SW271&instrument=i1%20DisplayPro%2C%20ColorMunki%20Display%2C%20Spyder4&html=1

But not sure which is correct or how to tell as they are certainly different.

3 – Or is it recommended to forget the HW Profile completely and start from scratch on a reset User/Custom monitor setting or something?
Which is where I read somethings about possible issues with Mac, Adobe and banding with Nvidia or Radeon, etc. with Graphic Card luts or something along those lines.

I think these are the main questions to get me started down the right path but may have an additional question depending on the answers.

Thanks again and I sent you guys a donation last night for your help and awesome app!
Geo

• This topic was modified 4 years, 7 months ago by Gette.

Calibrite Display Pro HL on Amazon  
Disclosure: As an Amazon Associate I earn from qualifying purchases.

[Vincent]

PME uses RGBLED spectral correction. It is not the proper one. Proper one could be a QLED, a WLED PFS (AdobeRGB flavor, Z24x) or less likely an old GBLED.

You can use (i) icon on correction to see spectral power distribution plot. For example:
https://colorimetercorrections.displaycal.net/hash/d24dda8b1aeb72bf9acb76a84efffc03/BenQ%20SW271%20%28i1%20Pro%202%29.ccss
is a GBLED
The other ones look like GBLED too*. Try it by youself using DisplayCAL (i) icon near spectral correction combo box or argyllcms specplot command.

So try PME with you desired target (native gamut D65 2.2 gamam looks like a good starting point), then test it with DisplayCAL (user made from community or standard GB-LED CCSS). PME ICCv2 profilers are not ok, do not woerry seesing high perofile white vs measured white. Just take a look on asumed white vs measured white and look for huge oscilations in a*=0 axis for greys.
If that is OK you can trust that HW cal.

(*) some people vandalized community corrections reagarding Benq monitors, to make them look like if they were using other technologies, for example:
https://colorimetercorrections.displaycal.net/?get&type=ccss&manufacturer_id=GSM&display=31MU97&instrument=i1%20DisplayPro%2C%20ColorMunki%20Display%2C%20Spyder4&html=1
But DisplayCAL server was smart enough to classify spectral “CLONES” up to 8th decimal place, or vandals were stupid enough to do not edit LG 31MU metadata.
So I would say that SW271 looks like a GBLED at 99% and vandalization seems unlikely… but these days with all these “ambassadors” making spam blogs for “reviewing” without actual data (uniformity, grey neutrality & gray range oscillations), you can have your doubts.

• This reply was modified 4 years, 7 months ago by Vincent.

[Vincent]

PME is using RGBLED cporrection so you can use it too to validate with DisplayCAL and check if following its own rules (correction for RAW measurements) PME did a good job: whiteness (measured vs assumed) & grey range (a*=0).

At the same time you can validate using GB-LED CCSS to check if it is close to “what it should be”. Well behaved i1d3 colorimeters should get close <2dE readings fro white using GBLED or RGBLED on a GB-LED dislay.

[Gette]

Vincent,

Thank you so much.

I will give it a shot. Quick initial question:

I do not see the “standard GB-LED CCSS” you mentioned. I also checked the actual file names in Finder.

If I were to guess it would be, but I don’t want to guess lol:
“LCD GB-r-LED IPS (Dell U2413)” or
“LCD GB-r-LED/RG Phosphor LED Family (AUO B156HO1 V.4, Dell u2413)”
with the later having a lot more curves (is that better or worse?).

As I assume a professional 1nm version(s) would be better the any of the 3nm user versions.

Attached is what I have.

Thanks!
Geo

Attachments:

You must be logged in to view attached files.

[Vincent]

• RG_Phosphor_Family_25Jul12.ccss is a U2413 GBLED plus other non GB-LED backlights.
• Dell_U2413_25Jul12.ccss is U2413 GB-LED standalone.

There is a slight different GB-LED sample in NEC software (PA242W) and that EDR can be translated to text file CCSS format. It’s a GB-LED but with sligthly shorter wavelength in blue led spike.
I do not remember peak position in blue spike in user made SW271 samples, but since they are 10nm it is difficult to take them as a reference for choosing the best 1nm vendor correction (U2413 vs PA242W).

So, choose U2413 standalone: “LCD GB-r-LED IPS (Dell U2413)”

[Gette]

Thank you Vincent.

I tried them both and not having luck with anything in trying to get neutral grays.

I tried various HW calibration and profiles in  PWE.

I tried HW calibration with DisplayCal profiles.

I even tried DisplayCal profiles on factory HW Calibrations (AdobeRGB).

By far the best is factory HW calibration and the stock/default (which I believe is just a pass through) ICC profile.
Grays are perfect. But White Balance 7000k and Brightness 130 are too high.

I’m thinking I am going to have to abandoned the 4k AdobeRGB 27-32″ screen as their doesn’t seem to be a decent one available around under $2k.

I think I am going to order a NEC PA271Q now. lol

Thank you!
Geo

[Vincent]

It is possible that Benq (&others) HW cal software is not able to get neutral greys if uncalibrated gray ramp is bad. Too few measurements. HW itself should be able since all of these at least have 1024 entry 1DLUT (+ matrix or + full 3DLUT for gamut emulation), thats why factory calibration looks smooth.

DisplayCAL GPU calibration is able to get smooth “HW cal”-like calibrations with an acurate device like an i1d3 if GPU+driver is suitable for this task and you choose slow (or medium) speed: Radeons (any config with propietary driver in Windows) or nvidias if connected by DP/HDMI with >8bpc seleted in driver (windows).
Check nvidia 1070 driver and set 10bpc in Displayport if ou did not test this way, then realibrate in GPU with slow speed. For example try the full GPU approach in Custom/User OSD mode with access to RGB gains.

[Gette]

Vincent,

Thanks again.

I have a Mac and currently have the Nvidia 1070 installed as I believe it is still better for Photoshop over my RX 580.

Also I have been doing Single Curve + Matrix as recommended due to Mac.

Thanks again!
Geo

PS  – Any issues with NEC PA271Q? I haven’t seen any so I assume their SpectraView II software and profiles are correct for that monitor (Panasonic panel I belive).

[Vincent]

 

PS  – Any issues with NEC PA271Q? I haven’t seen any so I assume their SpectraView II software and profiles are correct for that monitor (Panasonic panel I belive).

There ase some issues in some PA271Q units when using NEC SV2 software and no solution has been found (by user 2 user) so far other than return to store and try a different unit.

There are some webs online with extensive user2user report of those malfunctioning units because malfunctioning software, or firmware or HW+FW Spectraview engine inside those faulty PA271Q: LuLa or some polish forums.

Are there some distribution channels  (countries) more affected than others? IDNK, but not all PA271Q are affected and no FW update seems to solve this for the bad units, so IMHO it points to some issues in Spectarview Engine. Not related to panel, but all calibration stuff stored in monitor.

If you can try & return, if it is faulty it shloud fail at 1st try with SV2 self validation of resulting grey ramp calibration. But you should know that SV2 is not free, you have to pay for it although there are some monitor + software bundles. Make sure you can return to store these bundles before buying becasue software cannot be returned.

CS2730/CS2731 should be cheaper but 1000:1 and no LUT3D for emulating other devces behavior, but AFAIK ColorNavigator7 has no issues so far and it’s free.

[Author]

Posts